Longitude by Dava Sobel.
I'll tell you what I hate, right - there you are, an early-18th-century ship's captain and supposedly competent navigator, leading a crew of experienced seamen (no, stop it) and supposedly competent navigators, confidently under full sail returning from Gibraltar to London; you've just sat back and lit up a pipe when the entire fleet dashes itself to pieces on some rocks you didn't even know were there. A bit embarrassing for you, or at least it would be if you hadn't been drowned along with a couple of thousand of your men, or, possibly, depending on which version of the legend you believe, nearly drowned, washed up on a remote shore and then promptly murdered for some of the shiny trinkets on your person.
Anyway, this was far from an isolated incident and by the early 18th century the Navy and various other bodies overseeing military and commercial sea travel were getting pretty sick of it. The principal problem they identified was an inability to determine longitude with an acceptable degree of accuracy - the sort of degree of accuracy that prevents you accidentally crashing into a collection of islands and jagged bits of rock several miles wide.
This is not the place for a detailed lecture on maritime navigation, nor do I have the expertise or inclination to deliver one, but the main point is that to know your position on the earth's surface you need two co-ordinates, latitude and longitude; latitude is pretty easy to calculate and longitude is surprisingly hard in comparison. Many methods were used over the years, from the scientifically-sound but unbelievably complicated in practice such as observing the transits of the various satellites of Jupiter, to the simpler but also bonkers such as stabbing a dog and then dipping the knife in some magic powder at set intervals.
The simplest solution is just to have two clocks, one set to the time of whatever fixed point you're using as a meridian (nowadays this would be Greenwich but there have been a a few rival ones in use over the centuries, notably Paris) and one set to whatever the local time is at your location. It is a simple matter to then convert this time difference into a longitude. The bit that is far from simple is to produce a clock that is accurate enough for a long enough period of time and in a variety of challenging conditions - wide fluctuations in temperature and humidity, constant unpredictable motion, occasionally being eaten by a giant squid - something that up to the 18th century had been beyond even the most skilful clockmakers.
But *voice-over-guy voice* into this world of approximate timekeeping one man must walk alone, and that man's name is ... John Fashanu. No, wait *checks notes* John Harrison. A carpenter rather than a clockmaker by trade, he dabbles with clocks in his spare time, eventually moving on to making them, originally with the workings being made primarily of wood. Intrigued by the public challenge issued by the Longitude Board (and perhaps a bit by the generous financial reward for a workable solution), he starts working on clocks specifically designed to work at sea, first some large and spectacular steampunk-looking beasts and then, eventually, after more than twenty years of work, the pocket-watch-sized H4, in the process of all this work casually inventing a whole host of technological and engineering innovations like the bimetallic strip.
Despite H4 and some of its larger predecessors performing well in excess of the accuracy requirements set out by the Board of Longitude, Harrison and his son William never received the full sum promised by the terms of the challenge - this is partly down to Harrison's own perfectionism and the great length of time it took to produce the prototypes, but also perhaps some skullduggery by those with something to gain from promoting the other astronomical methods, in particular Harrison's nemesis and Astronomer Royal, Nevil Maskelyne.
Of course a good story requires heroes and villains, and Sobel makes no secret of Harrison being the hero here, plucky no-nonsense Northern lad pitted against the pompous stuffed shirts at Navigation Central and coming out on top thanks to pluck and grit. The role of villain is played by Nevil Maskelyne (more like Evil Maskelyne, amirite), wielding his royally-appointed power to promote his own impractical methods and doing his best to discredit the results of the marine chronometer tests and even "accidentally" dropping bits of the mechanism on the floor in the course of "inspecting" them.
I daresay Maskelyne was a more nuanced character than the one portrayed here, but a bit of a broad-brush approach is inevitable in a popular science book, and moreover a pretty short one - my Harper Perennial paperback edition is a large-print 175 pages with quite a lot of chapter breaks. I suppose I'd have preferred a bit more meat on the bones, as it's an interesting story, but it gets the job done pretty effectively. Longitude won several science and general non-fiction awards when it was published in 1996, and was made into a TV series starring lots of famous names including Michael Gambon as Harrison, where I'd imagine further liberties were taken with historical facts in order to provide a bit of narrative drive.





















